In India an important policy initiative has been the devolution of financial responsibilities to village level local governments called the Panchayats. The Preamble to this initiative is two fold. First such devolution would not only lead to increased public expenditure but also such expenditures being targeted in a manner consistent with the preferences and needs of the local population. Second, the local tax base would widen, thereby reducing the magnitude of the equalization transfers. However, the incentive structures behind the granting of such additional financial powers have been inadequately articulated. The results have been in the form of reduction in taxes collected, as well as a perceived shrinking of the tax base. These outcomes are posited by us to be due to ignoring the impact of cost of collecting taxes, as well as perverse impacts of devolution of expenditure decisions on local wages and profits. The extant literature has been so far unable to adequately explain the perverse outcomes of devolution especially where reactions to local tax efforts to transfers from the higher level governments are concerned. This paper has attempted to fill this gap. It models and measures the cost of taxation and uses this and the ratio of transfers that augment the local wage rate to those that do not, after controlling for a number of other village level characteristics, to explain tax collected at the local level within a framework that allows for mutual endogeneity of tax collected and transfers. We find that both the cost of tax collection and the ratio of transfers that augment the local wage rate to those that do not have a significant negative effect on tax collection, thus validating the conclusions of the theoretical model developed in this paper. Several policy conclusions are derived.